Multifunctional serine protease inhibitor-coated water-soluble gold

Subscriber access provided by READING UNIV

Article

Multifunctional serine protease inhibitor-coated watersoluble gold nanoparticles as a novel targeted approach for the treatment of inflammatory skin diseases David Limon, Maria Jose Fabrega, Ana C. Calpena, Josefa Badía, Laura Baldoma, and Lluïsa Pérez-García Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.7b00717 • Publication Date (Web): 06 Feb 2018 Downloaded from http://pubs.acs.org on February 10, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Bioconjugate Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Bioconjugate Chemistry

Multifunctional serine protease inhibitor-coated water-soluble gold nanoparticles as a novel targeted approach for the treatment of inflammatory skin diseases David Limón, †,ǂ María José Fábrega,§,ǁ Ana C. Calpena,ǂ,| Josefa Badia,§,ǁ,* Laura Baldomà,§,‖ and Lluïsa Pérez-García†,ǂ,*,# †

Departament de Farmacologia, Toxicologia i Química Terapèutica, Universitat de Barcelona,

Av. Joan XXIII, 27-31, 08028 Barcelona, Spain. ǂ

Institut de Nanociència i Nanotecnologia (IN2UB), Universitat de Barcelona, Av. Joan XXIII,

27-31, 08028 Barcelona, Spain §

Secció de Bioquímica i Biologia Molecular, Departament de Bioquímica i Fisiologia, Facultat

de Farmàcia i Ciències de l’Alimentació, Universitat de Barcelona, Av. Joan XXIII, 27-31, 08028 Barcelona, Spain. ǁ

Institut de Biomedicina de la Universitat de Barcelona (IBUB); Institut de Recerca Sant Joan de

Déu (IR-SJD), Barcelona, Spain. |

Departament de Farmàcia, Tecnologia Farmacèutica i Fisicoquímica, Universitat de Barcelona,

Av. Joan XXIII, 27-31, 08028 Barcelona, Spain.

ACS Paragon Plus Environment

1

Bioconjugate Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 48

KEYWORDS: Serine protease inhibitors, Water soluble gold nanoparticles; Skin diseases; Drug delivery; Kallikrein; Keratinocytes; Rosacea.

ABSTRACT: The overexpression and increased activity of the serine protease Kallikrein 5 (KLK5) is characteristic of inflammatory skin diseases such as Rosacea. The use of inhibitors of this enzyme - such as 4-(2-aminoethyl)benzenesulfonyl fluoride hydrochloride (AEBSF·HCl) or the anti-human recombinant Kallikrein 5 (anti-KLK5) antibody - in the treatment of the disease has been limited due to their low bioavailability, for which their immobilization in drug delivery agents can contribute to making serine protease inhibitors clinically useful. In this work, we synthesized gold nanoparticles (GNP) coated with a mixture of hydroxyl- and carboxyl terminated thiolates (GNP.OH/COOH), whose carboxyl groups were used to further functionalize the nanoparticles with the serine protease inhibitor AEBSF·HCl either electrostatically or covalently (GNP.COOH AEBSF and GNP.AEBSF, respectively), or with the anti-KLK5 antibody (GNP.antiKLK5). The synthesized and functionalized GNP were highly water soluble, and they were extensively characterized using UV-VIS absorption spectroscopy, Transmission Electron Microscopy (TEM), Dynamic Light Scattering (DLS) and Thermogravimetric Analysis (TGA). GNP.OH/COOH and their subsequent functionalizations effectively inhibited KLK5 in vitro. Internalization of fluorophore-coated GNP.OH/COOH in human keratinocytes (HaCaT cells) was proven using confocal fluorescence microscopy. Cell viability assays revealed that the cytotoxicity of free AEBSF is importantly decreased when it is incorporated in the nanoparticles, either ionically (GNP.COOH AEBSF), but most importantly, covalently (GNP.AEBSF). The functionalized nanoparticles GNP.AEBSF and GNP.antiKLK5 inhibited intracellular KLK5 activity in HaCaT cells and diminished secretion of IL-8 under inflammatory conditions triggered by TLR-2 ligands. This study points to the great potential of these GNP as a new intracellular


2

Page 3 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


delivery strategy for both small drugs and antibodies in the treatment of skin diseases such as Rosacea.

Introduction Rosacea is a chronic, inflammatory skin disease characterized by erythema, telangiectasia and papulopustules.1,2 Its etiology is not completely known yet and pharmaceutical products on the market commonly contain drugs that only act upon the medical symptoms.3–5 As the first line of defense against pathogens, keratinocytes express the Toll-like receptor TLR2 on their cell surface and in endosomal compartments, which can be activated by external agents.6,7 In Rosacea patients, the overexpression of TLR-2 enhances the sensitivity to external agents, increasing the expression and activity of the serine protease Kallikrein-5 (KLK5).8,9 In addition, the proteolytic activity of KLK5 is altered in this disease, producing pro-inflammatory cathelicidin peptides, which in turn leads to an increased IL-8 secretion.1,6–9 An overall biochemical pathway was drawn according to reports,1,8–11 and is represented in Figure 1. Enzymatic activity of KLK5 is regulated by endogenous serine protease inhibitors (SERPINS);12 however, in Rosacea patients, endogenous SERPINS seem not to have enough efficacy. The use of serine protease inhibitors as an attempt to treat Rosacea has been reported a few times,13–15 but the low bioavailability of some of them, such as AEBSF·HCl or the anti-KLK5 antibody, makes their conventional administration still limited. Therefore, the use of carriers in the administration of enzymatic inhibitors could help solving the problem for the regulation of KLK5 activity, helping to consolidate this new therapeutic strategy in Rosacea. Gold nanoparticles (GNP) are becoming a main interest as a platform for biomedical applications such as sensing and therapy, for their multiple advantages such as chemical stability, low toxicity,


3


Page 4 of 48

high extinction coefficient, and capability for varying the size and the external functionalizations according to the reaction conditions.16–18 According to the Enhanced Permeability and Retention Effect (EPR effect), the nanometric size of GNP promotes their selective extravasation from blood vessels to the inflamed tissues, where the permeability of vessels is higher,16,19,20 as is the case of the skin in patients with Rosacea.2,4,5 Therefore, GNP functionalized with the appropriate drugs can provide a passive targeting to the skin in Rosacea patients, decreasing the dose needed to get an effective treatment, as compared to the administration of the drug alone. Also, the functionalization with chemical entities which are capable of recognizing specific biomolecules, such as the case of antibodies, promotes retention of the particles around the tissues where their recognized antigens are located, providing an active targeting. For instance, the formation of a complex of GNP with an anti-KLK5 antibody could increase the concentration of such complex at the skin. Moreover, it is known that the size of GNP, as well as their coating agents, greatly influence their internalization in cells. For instance, GNP coated with PEG molecules can be more easily internalized in cells by increasing the length of the PEG chain and by decreasing the particle size.21 For these reasons, in this work PEG-coated gold nanoparticles were synthesized and functionalized with the well-known serine protease inhibitor AEBSF·HCl (covalently or electrostatically) or with the anti-KLK5 antibody. All these nanoparticles were extensively characterized, and their ability to be internalized in keratinocytes and to inhibit intracellular KLK5 and IL-8 secretion was assessed. The results suggest the potential of GNP to become an efficient new therapeutic approach for the efficient delivery of drugs for Rosacea treatment.


4

Page 5 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


B. oleronius D. folliculorum

External triggers •Corticosteroids •Hormones •UV radiation •Microorganisms (P. acnes)

TLR-2

Vit D 1,25 (OH)2 Transcription

Transcription

Inactive KLK 5

CAP 18 KLK 7

Proteolytic cleavage

Active KLK 5

Active KLK 7

Desquamation Rupture of corneodesmosomes

Cathelicidin peptides

• Cytokine secretion (IL8) •Angiogenesis •Chemotaxis •Erythema •Vasodilation •Thrombosis •Haemorrhage

Noninflammatory KR20 DI27 Inflammatory LL37 FA29

Figure 1. Biochemical pathway of Rosacea. The activation of TLR-2 receptors by external agents promotes by different routes the activation of KLK5. KLK5 activity is augmented and altered in Rosacea patients, leading to proinflammatory cathelicidins that orchestrate the inflammatory response. Adapted from 1,8–11

Results and discussion Synthesis and characterization of GNP.OH and GNP.OH/COOH gold nanoparticles With the aim of functionalizing gold nanoparticles with different serine protease inhibitors, our initial design was the synthesis of gold nanoparticles coated with one kind of thiolated-PEG molecules that include a terminal carboxyl group (PEG.COOH). The synthesized particles, GNP.COOH, showed an SPR band near λ= 515 nm using UV-VIS absorption spectroscopy. These particles remain in solution after centrifugation for 30 min, for which they can be considered


5


Page 6 of 48

highly-water soluble. However, further functionalization with serine protease inhibitors yielded nanoparticles with low water solubility, making them unsuitable for biological applications, and which were no longer used in further experiments. In order to obtain highly-water soluble particles, we considered using mixed coatings by including as well the thiolated PEG ligand with terminal hydroxyl groups (PEG.OH) as a water dispersing promoter. Thus, we synthesized nanoparticles GNP.OH/COOH, comprised of a gold core coated with a mixture of two different types of thiolated-PEG molecules: one with a hydroxyl terminal group (PEG.OH), which remains unreacted and provides water solubility, and a second one with a carboxyl terminal group (PEG.COOH), which can be further functionalized with serine protease inhibitors. Also, as control particles, hydroxyl-functionalized nanoparticles (GNP.OH) were synthesized, comprised of a gold core coated with only one kind thiolated-PEG molecules (PEG.OH) with a terminal hydroxyl group. A schematic representation of these nanoparticles is shown in Figure 2. Both GNP.OH and GNP.OH/COOH also show an SPR band near λ= 515 nm as observed by UV-VIS absorption spectroscopy (See Supporting Information, Figure S 1). Both GNP.OH and GNP.OH/COOH also remain in solution after centrifugation for 30 min, for which they can be considered highly-water soluble as well. For this reason, centrifugation-dissolution cycles could not be used for purification of GNP.OH or GNP.OH/COOH. Instead, other methods were used, such as dialysis or filter tubes for centrifugation. GNP were preserved in a stable solution at room temperature for at least one year with no apparent sedimentation or aggregation. As estimated by UV-VIS absorption spectroscopy,22 the GNP have an average diameter ≤3 nm, with an extinction coefficient of 1.49x106 L mol-1 cm-1 in both cases. Therefore, by considering the concentration of GNP in samples and the final volume of reaction, the number of thiols per


6

Page 7 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


nanoparticle was calculated. For instance, GNP.OH was estimated to be covered by 232 PEG.OH molecules per nanoparticle, and GNP.OH/COOH covered by 121 total thiol molecules (PEG.OH + PEG.COOH) per nanoparticle, with a proportion of PEG.COOH to PEG.OH of 1:1.3.

OH O O O

O

O

O

S

O

O

O

O

O

OH

O

OH

6 EG units

S

11 carbon chain

S

=

Au

Au

OH O O O

O

O

O

O

S

O

O

O

O

O

O

O

OH

6 EG units

S

OH

11 carbon chain Au

=

≤ 3 nm

Gold core

O

S S

OH

Au

~3.5 nm ~7.5-10 nm

Estimated size of ligand

Hydrodynamic diameter

Figure 2. Schematic representation of hydroxyl functionalized gold nanoparticles (GNP.OH) as control

(top), and mixed hydroxyl/carboxyl functionalized gold nanoparticles (GNP.OH/COOH) for further functionalization (bottom), in aqueous solution.


7


Page 8 of 48

Functionalization of gold nanoparticles GNP.OH/COOH were functionalized, using the carboxyl group from the particles surface, with different chemical entities such as the serine protease inhibitor AEBSF·HCl, linked both electrostatically (GNP.COOH AEBSF) (Figure 3, A) and covalently (GNP.AEBSF) (Figure 3, B), and the anti-KLK5 antibody covalently (GNP.antiKLK5) (Figure 3, C). All the different functionalized particles were highly water soluble, as they keep stable in solution after centrifugation for 30 min. O S

OH O

S

A

S

O

Au

O - +H

3N

GNP.COOH AEBSF O

OH O

S

B

F

S

S

F O

N H

Au

GNP.AEBSF OH O

S

C

S Au

HN

GNP.antiKLK5

Figure 3. Ionic and covalent functionalizations of GNP.OH/COOH. A) Ionic immobilization of AEBSF (GNP.COOH AEBSF). B) Covalent immobilization of AEBSF (GNP.AEBSF). C) Covalent immobilization of anti-KLK5 antibody (GNP.antiKLK5).

In the case of GNP.antiKLK5, the resulting functionalized particles are of great importance as they could play three important roles at the same time in the treatment of Rosacea. The first role is the passive targeting, achieved by the nanometric size of the platform, permitting a much higher


8

Page 9 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


extravasation in inflamed tissues such as the skin in Rosacea patients, than in healthy skin. The second role is the active targeting, which is provided by the antibody.18,23–25 As the anti-KLK5 antibody has a high affinity for KLK5 enzyme, their specific interaction could maximize the retention of the GNP at the tissues with higher concentration of KLK5, which is the case of the epidermis, especially in patients with Rosacea

1,9,26

. Finally, a third role is believed to be the

therapeutic potential of the the antiKLK5 itself, as it could possibly have a KLK5 inhibitory activity. Characterization of optimum GNP UV-VIS absorption spectroscopy All functionalized GNP (GNP.COOH AEBSF, GNP.AEBSF and GNP.antiKLK5) were characterized using UV-VIS absorption spectroscopy (See Supporting Information, Figure S 1). They all show the characteristic SPR band at ca. λ= 520 nm as well as bands corresponding to the serine protease inhibitors they contain.). Results suggest a successful functionalization in all cases, with no evidence of aggregation. Microscopy The size, morphology and polydispersity of GNP were confirmed by Transmission Electron Microscopy (TEM) (Figure 4), in GNP.OH (A) and GNP.OH/COOH (B) samples. GNP.AEBSF (C), GNP.antiKLK5 (D) and GNP.COOH AEBSF (E) were also observed to prove that functionalization does not induce particle aggregation.


9


GNP.OH

GNP.AEBSF

A

C

Page 10 of 48

GNP.OH/COOH

GNP.anti KLK5

D

B

GNP.COOH AEBSF

E

Figure 4. TEM images of different samples of gold nanoparticles. A) GNP.OH. B)

GNP.OH/COOH. C) GNP.AEBSF. D) GNP.antiKLK5. E) GNP.COOH AEBSF. Scale bar represents 100 nm.

For all samples observed, the size found for the majority of GNP was 1.5±0.7 nm, which also was the smallest size that could be detected due to the equipment resolution; the size distribution did not follow a typical Gaussian curve, but instead, they followed a right-tailed Fisher distribution (Figure 5). Histograms show that the different GNP are monodispersed, as more than 40% of the counts have a diameter around 1.5 nm. However, the polydispersity of GNP.OH seems to be slightly higher than the rest, as the percentage of particles above 4 nm in diameter is 25%, while in the rest of the samples, the percentage above 4 nm is less than 15%. This higher polydispersity can readily be observed in the plot as the slope of the curve is lower for GNP.OH than for the rest of the particles. Size distribution plots from the different GNP shown as both total counts and relative count (%) can be seen in Supporting Information (Figure S 2).


10

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Relative counts (%)

Page 11 of 48

Comparative view

100%

GNP.OH

80%

GNP.OH/COOH GNP.COOH AEBSF

60%

GNP.AEBSF GNP.antiKLK5

40% 20% 0% 0

1

2

3 4 5 6 Diameter (nm)

7

8

9

Figure 5. Size distribution of samples of gold nanoparticles GNP.OH (n=199), GNP.OH/COOH (n=583), GNP.AEBSF (n=2809), GNP.antiKLK5 (n=811), and GNP.COOH AEBSF (n=1778), expressed as a comparative view in relative counts (%).

LDI-TOF mass spectrometry The successful covalent functionalization was proven by Laser Desorption Ionization – Time of Flight Mass Spectrometry (LDI-TOF MS), by analyzing both GNP.COOH AEBSF and GNP.AEBSF in negative mode. For GNP.AEBSF, peaks observed at m/z 551 and 566, correspond to the AEBSF molecule covalently bonded through an amide linker to the polyethylene glycol spacer, fragments including six ethylene glycol units and either a two- or three-carbon chain respectively. These results show the effectiveness of the strategy for covalent functionalization. For GNP.COOH AEBSF, peaks observed at m/z 551 and 200 and 159 correspond to fragments of AEBSF molecule, while peaks observed at m/z 177 and 103 correspond to carboxyl-terminated


11


Page 12 of 48

ethylene glycol fragments. As expected, no peaks indicating a covalent functionalization of AEBSF were found (See Supporting Information, Figure S 3 and Figure S 4). Thermogravimetric analysis (TGA) Thermogravimetric analyses were performed with GNP.OH/COOH, GNP.AEBSF, GNP.antiKLK5, GNP.COOH AEBSF and AEBSF·HCl in solution, observing mass degradations corresponding to the thiol/drug/antibody with which they are functionalized, and confirming the successful functionalization. The sample containing only the drug AEBSF·HCl in solution shows a mass loss between 200-400 ºC, corresponding to the drug degradation. In the case of GNP.AEBSF, significant mass losses were observed also between 200 – 400 ºC, corresponding to big amounts of drug in the sample. In the case of GNP.antiKLK5, the mass loss in this range of temperatures was much smaller, suggesting that the percentage of surface functionalized on these particles is much less than on those small-drug-functionalized particles (See Supporting Information, Figure S 5 -Figure S 9). The sample containing GNP.COOH AEBSF also shows a mass loss around 300 ºC, corresponding to the drug. In this case, the biggest mass loss observed around 100 ºC corresponds to high amounts of water, indicating the sample was not as dry as the rest of the particle samples. Therefore, the smaller mass loss at 300 ºC as compared to GNP.AEBSF does not mean a smaller proportion of drug in the sample, but only a higher water content. Thiolated-PEG molecules were not completely degraded at the temperatures reached (950 ºC), as seen by the lack of a mass stabilization of the sample, for which the total mass of them cannot be known. Dynamic Light Scattering (DLS) GNP.OH/COOH, GNP.COOH AEBSF, GNP.AEBSF and GNP.antiKLK5 nanoparticles were analyzed by DLS to assess their hydrodynamic diameters, their polydispersity, and whether


12

Page 13 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


they are aggregated or monodisperse in solution. The diameters observed are higher than the gold core diameters observed by TEM, as hydrodynamic diameter not only considers the core but also reflects the chemical functionalization. Hydrodynamic diameters observed are quite similar in all particles under study. GNP.OH/COOH showed the highest diameter average as also the highest dispersity (12.9±2.7 nm), suggesting these particles are aggregated in solution. The electrostatically functionalized particles GNP.COOH AEBSF showed the smallest hydrodynamic diameter (7.4±1.2 nm), while particles covalently functionalized with AEBSF (GNP.AEBSF) or with antibody (GNP.antiKLK5) showed an increase in their diameter (8.2±2.3 nm and 11.7±2.1 nm respectively). The sizes observed in the different samples can only be used as an estimation of the size and polydispersity of the particles, in accordance with the sensibility of the technique (See Supporting Information, Table S 1). Based on the results of the different techniques concerning particle sizes, a schematic representation of the nanoparticles, the different functionalizations, and their sizes is shown in the Supporting Information (Figure S 10).

In vitro inhibition of KLK5 As a first step, in vitro enzymatic assays were performed to test whether the different kinds of nanoparticles functionalized with serine protease inhibitors can inhibit KLK5 activity. Experiments were performed following the protocol described by the supplier.27 For activity assays, KLK5 was incubated with a specific fluorogenic substrate in the absence or presence of the

different

functionalized

nanoparticles

(GNP.COOH

AEBSF,

GNP.AEBSF,

GNP.antiKLK5), as well as with the free drug AEBSF·HCl or the anti-KLK5 antibody for comparison. GNP.OH and GNP.OH/COOH were used as controls. In Figure 6, the percentage of inhibition calculated for each type of drug/particles was plotted against the concentration of


13


Page 14 of 48

functional groups responsible for such inhibition (“active groups”). In GNP.OH/COOH, the term “active groups” refers to the concentration of PEG.COOH molecules. Likewise, it refers to those PEG.COOH molecules after functionalization in case of GNP.COOH AEBSF, GNP.AEBSF, and GNP.antiKLK5. Moreover, it represents the concentration of drug in case of GNP.COOH AEBSF, GNP.AEBSF and free AEBSF·HCl. As GNP.OH is only coated with one type of thiolate with hydroxyl terminated groups, concentration was adjusted to correctly represent an equivalent concentration of total thiolates than in GNP.OH/COOH taking in account the proportion of PEG.OH to PEG.COOH ligands. This adjustment was also taken into account in the rest of experiments. In case of GNP.antiKLK5 and anti-KLK5 in solution, the concentration of antiKLK5 antibody is also shown in the axis. A broad range of concentrations was chosen in order to determine whether the inhibition follows a certain model. The best fitting model was obtained for each kind of particles/drug/antibody and the IC50 values (the concentration at which 50% of enzymatic activity is obtained) were estimated. However, it is important to note that these in vitro conditions are expected to be different than those inside the cells, including the concentration of KLK5 enzyme; for which IC50 values are not expected to be the same in cells and are calculated only for comparison purposes.


14

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Enzymatic inhibition (%)

Page 15 of 48

Anti-KLK5 Ab (µg/mL)c

Active groups (µM)a 143

285

428

570

713

1140

1710 Anti-KLK5 Ab (pg/mL)b

Concentration Figure 6. In-vitro enzymatic inhibition of KLK5 in presence of different gold nanoparticles at different concentrations. aConcentration of active groups for GNP and AEBSF·HCl in solution (µM). bConcentration of anti-KLK5 antibody in GNP and in solution (pg/mL).cConcentration of anti-KLK5 antibody in solution (µg/mL). Values were expressed as Mean±SD (n=3). GNP.OH Followed a Boltzmann Sigmoidal model, while the rest of the particles and drugs followed a One Phase Exponential Association model

Results in Figure 6 show that the different particles/drug/antibody can actually inhibit KLK5 activity, although at different concentrations or following a different model. Equations, IC50 values and other inhibition parameters obtained can be seen in Supporting Information (Table S 2 and Table S 3).


15


Page 16 of 48

As expected, AEBSF·HCl in solution inhibits KLK5 activity, which could be attributed to the covalent attachment of the fluorosulphonyl group from the drug to the hydroxyl group of the serine in the enzyme’s active site, and thus being an irreversible inhibition. Interestingly, also control particles (GNP.OH and GNP.OH/COOH) display KLK5 inhibitory activity, suggesting a possible interaction between the GNP and the enzyme’s active site, for example, the formation of hydrogen bonds or ionic bonds. These interactions might therefore affect the normal enzymatic activity. GNP.OH showed a very different inhibitory behavior than the rest, following a Boltzmann Sigmoidal model, and providing much less inhibition as compared by the IC50 values (P0.05) were obtained between the inhibitory activity of the particles GNP.OH/COOH, GNP.AEBSF, GNP.COOH AEBSF, GNP.antiKLK5 or the free AEBSF.HCl; however, when the drug is ionically bonded in GNP.COOH AEBSF, a slight enhancement of the inhibition can be appreciated, which could be due to the non-covalent nature of the attachment, which permits some drug release, and therefore both the drug and the nanoparticles contribute to the overall inhibition. Statistical analysis can be seen in Supporting Information (Table S 4). Finally, the free anti-KLK5 did not show inhibitory activity at similar concentrations as when it is in GNP.antiKLK5 (50 – 1200 pg/mL). This can be explained as the molar concentration of KLK5 in the assay is almost one million times higher than the concentration of antibody. However, when the concentration of antibody is increased (1 – 20 µg/mL), inhibition is achieved, which also follows an exponential behavior. These results show that the interaction between the anti-KLK5


16

Page 17 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


antibody and the enzyme can actually inhibit the enzymatic activity, and which, to our knowledge, was not previously reported. Internalization of GNPs in human keratinocytes To test the ability of gold nanoparticles to be internalized in human keratinocytes (HaCaT), GNP.OH/COOH nanoparticles were labelled with the fluorophore BODIPY by covalently bonding the fluorophore to the carboxylic acid groups forming an amide bond. Labelled particles were added to keratinocytes at a concentration of 30 µM of active groups. After 24 hours of incubation at 37ºC, internalization of nanoparticles was visualized by confocal fluorescence microscopy (Figure 7). As expected, no red signal was observed in non-treated control cells. In contrast, BODIPY-labelled nanoparticles were visualized in the cytoplasm of HaCaT cells incubated at 37ºC. Remarkably, the area of the nucleus is observed completely dark, showing that the particles do not enter the cell nucleus, for which they should not induce damage to DNA. The cell internalization of drugs or proteins is hampered by their physicochemical properties; therefore, these particles could be used as an intracellular delivery system. For example, the potential use of anti-KLK5 antibody to block KLK5 in vivo, is hampered by the inability of antibodies to enter the target cells, as it is known that the cell membrane represents a nonpermissive barrier for the transport of antibodies across it. Therefore, as it is the case of GNP.antiKLK5, these nanoparticles could be used to drive cellular uptake of antibodies and target cytosolic KLK5 in HaCaT cells.


17


Control Membrane

HaCaT + GNP GNP

Page 18 of 48

Merged

Figure 7. Internalization of BODIPY-labeled GNP.OH/COOH in HaCaT keratinocytes by confocal fluorescece microscopy. HaCaT cells were incubated for 24 hours with 30 µM of GNP.OH/COOH. Images show cell membranes counterstained with wheat germ agglutiningAlexa-594 (green) and internalized BODIPY-labelled particles (red). Analysis was performed in a Leica TCS SP5 laser scanning confocal spectral microscope with 63x oil immersion objective lens, and images were captured with a Nikon color camera (16 bit). Scale bar: 20 µm.

Cytotoxicity assays Although it is generally accepted that the toxicity of gold nanoparticles is low because of the low reactivity of the material, there is a growing interest in the study of their toxicity and life expentancy. It has been seen that the size of the gold nanoparticles influences the cytotoxicity, being the ones with a gold core diameter smaller than 2 nm the ones that present unusual reactivity in terms of oxidative stress;28–30 however, toxicity is highly dependent on the outer functionalization of the GNP, and PEG entities relieve nanotoxicity and permit the GNP escaping the opsonization process in the immune response.31 Hence, the cellular toxicity in human keratinocytes (HaCaT cells) was evaluated after incubation with the different nanoparticles by performing the MTT assay.


18

Page 19 of 48 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Cell viability of HaCaT keratinocytes was evaluated both in presence or absence of the glucan related zymosan, with the aim of setting up a cellular model that mimics Rosacea conditions, and since it is known that zymosan can activate the TLR-2 receptors, whose signalling pathway is involved in Rosacea-associated overexpression of KLK5 (see Figure 1),6,7,9 The results show that zymosan shows no cytotoxicity after 24 h incubation. Moreover, a possible indirect influence of zymosan in cell viability after incubation with nanoparticles was evaluated by incubating HaCaT cells with GNP.OH/COOH, both in presence and absence of zymosan, showing also that zymosan has no influence on viability results (data not shown). For these reasons, the cell viability in human keratinocytes (HaCaT cells) was evaluated under Rosacea conditions (zymosan-stimulated) after incubation with the different nanoparticles at concentrations up to 100 µM (GNP.OH, GNP.OH/COOH, GNP.COOH AEBSF, GNP.AEBSF, and GNP.antiKLK5), as well as with the free drug AEBSF·HCl or the anti-KLK5 antibody (Figure 8). GNP concentrations were calculated on the basis of “active groups”. In GNP.OH/COOH the term “active groups” refers to the concentration of PEG.COOH molecules. Likewise, it refers to those PEG.COOH molecules that were functionalized in case of the other nanoparticles, namely GNP.COOH AEBSF, GNP.AEBSF, and GNP.antiKLK5. As GNP.OH does not have two types of thiols, concentration was adjusted to correctly represent an equivalent concentration of total thiols than in GNP.OH/COOH taking into account the proportion of PEG.OH to PEG.COOH thiols. GNP.OH did not cause toxicity in keratinocytes since cell viability was close to 100% at all the concentrations tested (Figure 8 A). These results were consistent with the reported alleviation of GNP toxicity by PEG functionalization. However, incubation with GNP.OH/COOH resulted in a significant decrease in cell viability, around 30% at concentrations above 5 µM (Figure 8 B). This effect could be attributed to the reactivity of free COOH groups on the surface of these


19


Page 20 of 48

nanoparticles, prior to the functionalization with the desired drugs. Despite the toxicity levels observed with these particles, they should not represent a problem when the administration route is intended to be dermal, as the replication of keratinocytes in the skin occurs faster than the majority of cells in the organism, and the outer layer of the skin (stratum corneum) is entirely comprised of biologically dead cells. In contrast, the serine protease inhibitor AEBSF·HCl was highly cytotoxic in solution, resulting in less than 10% of cell viability at 75 µM and undetectable cell viability levels at 100 µM, which completely impairs its potential therapeutic use. This high toxicity is most probably related to the reactivity of the fluorosulphonyl group, which might be interacting with other vital biomolecules of the cell. However, drug associated cytotoxicity was significantly reduced (p

Multifunctional serine protease inhibitor-coated water-soluble gold

Recommend Documents